DIGITAL SIGNAL PROCESSING ARCHITECTURE SUPPORTING EFFICIENT CODING OF MEMORY ACCESS INFORMATION

• US 20110314209A1
• Filed: 06/18/2010
• Published: 12/22/2011
• Est. Priority Date: 06/18/2010
• Status: Active Grant

• Current Assignee: Avago Technologies General IP PTE Limited (Broadcom, Inc.)
• Original Assignee: LSI Corporation (Broadcom, Inc.)
• Inventors: Eckstein, Erik
• Granted Patent: US 8,484,404 B2
• Time in Patent Office: Days
• US Class Current: 711/104
• CPC Class Codes: G06F 12/0623 for memory modules
• CPC Class Codes: G06F 9/355 Indexed addressing

Abstract

A digital signal processing architecture supporting efficient coding of memory access information is provided. In an example embodiment, a digital signal processor includes an adjustment value register to store an initial adjustment value and a succeeding adjustment value. The digital signal processor may also include an address generator circuit to retrieve an instruction including a memory address value that is greater than N, and a further instruction including a further memory address value that is less than or equal to N. In addition, the digital signal processor may include a memory, which includes a high bank address space defined by memory locations that are uniquely identified with memory address values greater than N. The address generator circuit may access the high bank address space, using initial adjustment value and the memory address value, or using the succeeding adjustment value and the further memory address value.

Claims

1. A digital signal processor, comprising:

an adjustment value register configured to store an initial adjustment value and a succeeding adjustment value;

an address generator circuit communicatively coupled with the adjustment value register and configured to retrieve an instruction including a memory address value that is greater than N and a further instruction including a further memory address value that is less than or equal to N; and

a memory communicatively coupled with the address generator circuit and including a high bank address space defined by memory locations that are uniquely identified with memory address values greater than N, the address generator circuit being further configured to

access the high bank address space, using the initial adjustment value and the memory address value, when the adjustment value register is not storing the succeeding adjustment value, and

access the high bank address space, using the succeeding adjustment value and the further memory address value, when the adjustment value register is storing a succeeding adjustment value.

2. The digital signal processor of claim 1, wherein the high bank address space is defined by memory locations that are uniquely identified with memory address values that reflect greater than 65,535 decimal units, the succeeding adjustment value reflects greater than the 65,535 decimal units, and the further memory address value reflects less than or equal to the 65,535 decimal units.

3. The digital signal processor of claim 1, wherein the address generator circuit is configured to process the succeeding adjustment value together with the further memory address value to calculate a binary sum of the succeeding adjustment value and the further memory address value.

4. The digital signal processor of claim 1, wherein the address generator circuit is configured to place, during start-up of the digital signal processor, a zero in the adjustment value register as the initial adjustment value, and to process the initial adjustment value together with the memory address value to calculate a binary sum of the initial adjustment value and the memory address value.

5. The digital signal processor of claim 1, wherein the address generator circuit is communicatively coupled to an instruction module, and responsive to the instruction module requesting storage of the succeeding adjustment value, the address generator circuit is configured to

confirm that the instruction module is permitted to cause a storage of the succeeding adjustment value, and

place the succeeding adjustment value in the adjustment value register when the address generator circuit has confirmed permission.

6. The digital signal processor of claim 5, wherein the instruction module includes a functional task of a machine code, and the further instruction is associated with the functional task.

7. The digital signal processor of claim 5, wherein the address generator circuit is configured to

place the succeeding adjustment value in the adjustment value register responsive to a request from an operating system to do so, the succeeding adjustment value to be used in association with a first functional task of a machine code, and the first functional task being associated with the further instruction,

place a superseding adjustment value in the adjustment value register responsive to a request from the operating system to do so, the superseding adjustment value to be used in association with a second functional task, the second functional task being associated with a yet a further instruction including yet a further memory address value that is less than or equal to N;

and

access the high bank address space, using the superseding adjustment value and yet the further memory address value, when the adjustment value register is storing the superseding adjustment value.

8. A method comprising:

setting an initial adjustment value;

in an absence of a setting of a succeeding adjustment value,

utilizing the initial adjustment value and a first storage location value that is greater than N to access a section of a storage; and

in a presence of the setting of the succeeding adjustment value,

utilizing the succeeding adjustment value and a second storage location value that is less than or equal to N to access the section of the storage, the section of the storage being defined by, and accessible with, unique storage location values that are greater than N and not less than or equal to N.

9. The method of claim 8, further comprising:

placing content of a unique storage location at a destination storage location,

wherein the utilizing of the succeeding adjustment value and the second storage location value includes processing the succeeding adjustment value together with the second storage location value to generate a unique storage location value of the unique storage location values, and wherein the accessing of the section of the storage includes accessing the section of the storage at a unique storage location identified by the unique storage location value.

10. The method of claim 8, wherein the succeeding adjustment value reflects greater than 65,535 decimal units, the second storage location value reflects less than or equal to the 65,535 decimal units, and the section of the storage is defined by storage locations that reflect greater than the 65,535 decimal units.

11. The method of claim 8, wherein the utilizing of the succeeding adjustment value and the second storage location value includes,

retrieving a binary succeeding adjustment value and a binary second storage location value, and

adding the retrieved binary succeeding adjustment value to the binary second storage location value.

12. The method of claim 8, wherein the setting of the initial adjustment value includes setting the initial adjustment value to zero, and the utilizing of the initial adjustment value and the first storage location value includes adding binary numbers representing each of the initial adjustment value and the first storage location value.

13. The method of claim 8, wherein the setting of the succeeding adjustment value includes setting the succeeding adjustment value in response to a request from an instruction module, and the utilizing of the succeeding adjustment value and the second storage location value includes,

extracting the second storage location value from a memory access instruction, and

processing the succeeding adjustment value together with the second storage location value responsive the memory access instruction.

14. The method of claim 13, wherein the setting of the succeeding adjustment value includes,

verifying that the instruction module is authorized to activate the setting of the succeeding adjustment value, and

setting the succeeding adjustment value if authorization of the instruction module is verified.

15. The method of claim 13, wherein the setting of the succeeding adjustment value in response to the request from the instruction module includes setting the succeeding adjustment value in response to the request from a functional task of a machine code, and wherein the memory access instruction is associated with the functional task.

16. The method of claim 13, further comprising:

setting a superseding adjustment value; and

accessing a different section of the storage via utilizing a superseding adjustment value and a third storage location value that is greater than N, the different section of the storage being defined by different unique storage location values that are greater than N and not less than or equal to N,

wherein an operating system requests the setting of the succeeding adjustment value, the memory access instruction is associated with a functional task of a machine code, and an additional memory access instruction that includes the third storage location value is associated with an additional functional task of the machine code.

17. A machine-readable medium including instructions that when executed by a machine, cause the machine to perform the method of claim 8.

18. A networked digital media system comprising:

an input/output module configured to receive digital media input via a network and output digital media to a user;

a random access memory (RAM) having a high bank address space defined by memory locations uniquely identified with memory address values greater than N; and

a digital signal processor communicatively coupled to the RAM and including,

an adjustment value register configured to store an initial adjustment value and a succeeding adjustment value,

an address generator circuit communicatively coupled with the adjustment value register and being configured to retrieve an instruction including a RAM address value that is greater than N, and a further instruction including a further RAM address value that is less than or equal to N, and the address generator circuit being further configured to

access the high bank address space, using the initial adjustment value and the RAM address value, when the adjustment value register is not storing the succeeding adjustment value, and

access the high bank address space, using the succeeding adjustment value and the further RAM address value, when the adjustment value register is storing a succeeding adjustment value.

19. The networked digital media system of claim 18, wherein the networked digital media system is selected from the group of devices consisting of a mobile phone, a mobile video device, a mobile audio device, and a mobile audio/visual device.

20. The networked digital media system of claim 18, wherein the high bank address space only includes memory locations that are uniquely identified with memory address values that reflect greater than 65535 decimal units, the succeeding adjustment value reflects greater than the 65535 decimal units, and the further RAM address value reflects less than or equal to the 65535 decimal units.

21. The networked digital media system of claim 18, wherein the digital signal processor includes a plurality of adjustment value registers and a plurality of address generator circuits, the adjustment value register being one of the plurality of adjustment value registers, and the address generator circuit being one of the plurality of address generator circuits, and wherein the plurality of adjustment value registers and the plurality of address generator circuits have substantially similar configurations to the adjustment value register and the address generator circuit.

22. A method comprising:

requesting that a specific adjustment value be set in a processor register;

instructing a processor to access a high bank of memory using a low bank memory address that is less than or equal to N; and

confirming that the processor has accessed the high bank of memory at a high bank address,

the high bank address being based a sum of the specific adjustment value and the low bank memory address, and the high bank of memory being defined by memory addresses that are greater than N and not less than or equal to N.